About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Lilly Shrinks | Main | The Drug Business: A Turbulent Future? »

September 16, 2009

Real Electrons

Email This Entry

Posted by Derek

I posted images of real pentacene molecules the other day, but now the single molecule/single-atom imaging field has reached another milestone. There's a paper coming out in Physical Review B from a team in Kharkov using a field emission electron microscope. At heart, that's a pretty old type of machine, first invented back in the 1930s, and it's long provided images of the arrangements of atoms in metal surfaces. (More precisely, you're getting an image of the work function, the energy needed to remove electrons from the material).

But this latest work is something else entirely. The researchers have improved the resolution and sensitivity, narrowing things down to single-atom tips. So instead of a tungsten surface, we have a single carbon atom at the end of a chain. And instead of the behavior of the electons in a bulk metal, we have the electron density around one nucleus. Behold the s and p orbitals. Generations of students have learned these as abstractions, diagrams on a page. I never thought I'd see them, and I never thought I'd see the day when when it was even possible. As always, I react to these things with interest, excitement, and a tiny bit of terror at seeing something that I assumed would always be hidden.

Comments (53) + TrackBacks (0) | Category: General Scientific News


1. anon the II on September 16, 2009 8:45 AM writes...

I'll readily admit that I don't understand this stuff very well. But I'm having a hard time understanding how you can get this carbon atom to lie on it's side so that you can see one well resolved p-orbital. Shouldn't there be one perpendicular at the same time? It would be isoenergetic. I'm not so sure you could get this image if you didn't have some idea of what it might look like.

Permalink to Comment

2. Sili on September 16, 2009 8:46 AM writes...

Allow me to repeat the common gripe about the Nature cover of a decade back.

These are electron densities the square of the orbital. You cannot see an orbital. It's a physical impossibility.

Permalink to Comment

3. Anonymous on September 16, 2009 8:47 AM writes...

unreal ... awe inspiring

Permalink to Comment

4. Anonymous on September 16, 2009 9:10 AM writes...

Shouldn't this be in nature/science as well?!

Permalink to Comment

5. Elizabeth on September 16, 2009 9:11 AM writes...

Oh my goodness.

That may be one of the most awe-inspiring things ever.

I can't believe we can SEE them.

Permalink to Comment

6. Cyclase on September 16, 2009 9:54 AM writes...

Although I'm not a physical chemist, being a phys-chem lover, I have a gut feeling that this paper is bogus... I've yet to read the paper, but I do have the same concerns as "anon the II", and I do feel an overall sense of uneasiness rather than awe when I see these images.

Permalink to Comment

7. Rhenium on September 16, 2009 10:20 AM writes...

#4 There was a similar paper in Science a while back, I'll try digging the reference up later.

Permalink to Comment

8. nitric_oxide_99 on September 16, 2009 10:46 AM writes...

God never saw an orbital.

-Walter Kauzmann to Maitland Jones

Permalink to Comment

9. GA on September 16, 2009 11:57 AM writes...

Is it my eyes or is the s-orbital not perfectly circular in this projection?

Permalink to Comment

10. DrSnowboard on September 16, 2009 12:21 PM writes...

Is it me, but if you peer really hard at the s orbital, you can see the silhouette of Barack Obama...?

Permalink to Comment

11. Aspirin on September 16, 2009 1:46 PM writes...

So this is being published in Phys Rev while the other stuff was in Science. The other stuff was definitely Science-worthy but so is this (if it really pans out that is). While Phys Rev is a great journal, does someone smell at least some politics here? IBM clout and all that.

Permalink to Comment

12. Chemoptoplex on September 16, 2009 1:47 PM writes...

If it were in a chain wouldn't the p-orbital be hybridized or involved in some MO construct? Why do we see a perfect p-orbital with its hair combed posing for the camera?

Permalink to Comment

13. Paul S on September 16, 2009 2:36 PM writes...

Holy catfish.

I'm with some of the skeptical posters here. I'd like to see this reproduced.

Permalink to Comment

14. Lucifer on September 16, 2009 2:51 PM writes...


You could give the devil his due, once in a while.

Permalink to Comment

15. Curious Wavefunction on September 16, 2009 2:53 PM writes...

The problem with such images is that there is inevitably some image enhancement involved; it's not like you see an electron cloud the way you would see a pollen grain under a light microscope. This always makes it hard to judge how "real" such data is. I would defer to the experts to tell us the limitations of such a study.

Permalink to Comment

16. Lucifer on September 16, 2009 3:03 PM writes...

Curious Wavefunction,

Do you think images from the hubble telescope have any resemblance to unprocessed images of those astronomical objects? or protein backbones look like stylized coils or sheets? or all hydrogen bonds look like dashed lines.

What is more important is our newfound ability to directly measure the phenomena at that level and confidence.

Permalink to Comment

17. milkshake on September 16, 2009 3:13 PM writes...

It was fotoshopped - no way it is real orbitals (I think it is actually butt-prints)

Permalink to Comment

18. Lucifer on September 16, 2009 3:24 PM writes...


It seem that they have been dabbling in this area for a few years now..

Field-ion microscopy of quantum oscillations of linear carbon atomic chains.(2009)

Imaging of Atomic Orbitals: Quantum End States in Carbon Atomic Chains.(2009)

Low-temperature field ion microscopy of carbon nanotubes.(2007)

Preparation and characterization of monoatomic C-chains: unraveling and field emission. (2007)

Permalink to Comment

19. Steve on September 16, 2009 3:41 PM writes...

This is odd. I've always pictured that discrete s, p, d, and f-orbitals and hybridized orbitals exist, but wouldn't they overlap? How could you see an s-orbital and a p-orbital separately?Aren't orbitals actually spatial probabilities of where an electron of a particular energy and spin would exist in the space around a nucleus? What about Pauling's hybrid orbital theory? If this were a terminal carbon atom in an organic molecule, wouldn't it have 4 sp3 hybrid orbitals instead of discrete s and p orbitals? To "see" the p-orbital, would the carbon atom in view have to be a radical with a singly occupied p-orbital? If it's a carbocation with an empty p-orbital, could you still "see" it? How do I explain the implications of these research findings to students when I teach organic chemistry?

Permalink to Comment

20. Kicker on September 16, 2009 4:10 PM writes...

It's great to know that electrons are blue. I say hunt down the PIs and take back the money.

This is stupid science.

Permalink to Comment

21. Lucifer on September 16, 2009 4:18 PM writes...

Have a look at linked article from the same group, as it tells you a lot about their techniques. Look at figure 3 and 5.

Permalink to Comment

22. Hap on September 16, 2009 4:31 PM writes...

#12 brought it up, but why do the areas of squared electron density look like those for the individual s and p orbitals rather than those for the hybridized orbitals? I know (at least vaguely) that the hybridization of orbitals is imaginary, a way of generating orbitals with appropriate properties to form the bonds whose behavior we observe, but then shouldn't we be seeing electron density correlating to an orbital with those properties rather than the electron densities of the individual atomic orbitals? I guess I'd like to have an idea exactly how far out of my depth I am here.

Permalink to Comment

23. Broklynite on September 16, 2009 5:32 PM writes...

I recall seeing images of discrete s and p orbitals around ten years back, but only vaguely recall the details (I was more in shock that it happened).

That said- Kicker, if it hadn't been for your third sentence, I would have thought you were kidding. Color is added to these kinds of images in order to highlight, differentiate, emphasize or dramaticize. I admit that I have not read the article. That said, I would heavily suspect that the claim is not being made that the orbitals are blue, and it would be understood by most people that this would not be the case.

Again, I haven't read the article. They might very well claim that they are blue. But I suspect not.

Permalink to Comment

24. Tom on September 16, 2009 7:37 PM writes...

These aren't the same guys who discovered "poly-water" are they?

Permalink to Comment

25. Jose on September 16, 2009 8:12 PM writes...

How does one see a probability surface- shouldn't the edges of the orbitals slowly fade out into infinity? Textbook diagrams have some sort of a CI build in, but how can a real atom? Perhaps this tailing gets lost in the S/N? Any experts out there?

Permalink to Comment

26. Curious Wavefunction on September 16, 2009 9:46 PM writes...

-Do you think images from the hubble telescope have any resemblance to unprocessed images of those astronomical objects?
I would think it would be much more tricky to get images of subatomic domains than domains that are light years across.

As for protein backbones, there is certainly some uncertainty in the atomic positions. But this uncertainty is usually well-documented. Plus, whatever data there is can be cross-checked by independent methods thus adding to its reliability (or lack thereof).

All I am asking is whether all possible artifacts inherent in such a study have been accounted for. The paper does not seem to be out yet.

Permalink to Comment

27. Bored on September 16, 2009 10:10 PM writes...

The skepticism over a couple of little pictures here is refreshing. Now if we could just apply some of that skepticism to anthropogenic global warming........

Permalink to Comment

28. jewbag on September 17, 2009 3:32 AM writes...

global warming is made by elves and all of their damned cookies

Permalink to Comment

29. anon the 3rd on September 17, 2009 4:36 AM writes...

@anon the ii

Its could be due to electron spin, +/- 1/2. Basically the atom may act like a mini gyroscope.

Permalink to Comment

30. Anonymous on September 17, 2009 8:05 AM writes...

Hello? "Skeptic" doesn't mean "naysayer". A skeptic is someone who investigates and based on this information, can issue more than just half-assed feel-so opinions.

The claim is very simple: they have brought the tip temperature down to such a state that single quantized states can be observed. Sounds very much like the American work to produce a Bose-Einstein condensate.

Permalink to Comment

31. Vette0804 on September 17, 2009 9:23 AM writes...

Why do we not see an s-orbital along with the p-orbital?

Permalink to Comment

32. Anonymous on September 17, 2009 9:25 AM writes...

"I'm not so sure you could get this image if you didn't have some idea of what it might look like."


Permalink to Comment

33. Aaron on September 17, 2009 9:39 AM writes...

@bored -- Thousands of scientists have indeed been applying the same skepticism you see here to the thousands of papers which have been published on climate change in the last few years. There is still intense debate about the role of things like the thermohaline circulation and changes in cloud cover.

Over the years, however, an overwhelming consensus has emerged that CO2 concentrations correlate to global temperature in the quaternary. I have seen no scientist even try to argue that measurements taken all over the world that show an increase from about 300 parts per million CO2 to 370 ppm are insignificant or incorrect, nor that human release of CO2 is the primary factor in that increase.

This abstract about the orbitals, however, reeks of pseudoscience. Should be quite obvious once the paper is published (if it ever is, I have serious doubts.)

Permalink to Comment

34. Jose on September 17, 2009 10:12 AM writes...

Since solvated electrons are beautiful cobalt blue, real orbital densities must be the same color, right?

Permalink to Comment

35. waywardchemist on September 17, 2009 10:39 AM writes...

@Steve, Hap and everyone asking about not seeing hybridization -- Hybridization is a somewhat useful model for generating geometry but it does a terrible job on orbital energies/shapes. The best experimental evidence of this comes from photoemission data of methane (done back in the late 60s early 70s, probably by U. Gelius or P. Siegbahn). If hybridization were true, we would expect a single feature in the valence band corresponding to the four isoenergetic sp3 orbitals. Experimentally, we observe two features several eV apart with a ratio of 1:3 (ignoring some technical details such as spin orbit splitting and vibrational features for simplicity) The features are assigned as 2a1 and 1t2 by MO theory. Each of the isoenergetic C-H MOs is composed of partial overlap between the H 1s and both C 2s and 2p orbitals but the contributions from the C 2s and 2p electrons are still energetically distinguishable.

Without spending too much time looking into it, the data presented look slightly distorted from what is expected for atomic orbitals. My guess would be that distortion is from the bonds that have been formed.

Permalink to Comment

36. Hap on September 17, 2009 10:45 AM writes...

Thanks for the explanation. I suck at seeing MOs.

Permalink to Comment

37. Brok on September 17, 2009 11:32 AM writes...

Ur...much as I enjoy a good debate on global warming, perhaps this isn't the place for it...?

Permalink to Comment

38. bored on September 17, 2009 4:58 PM writes...

#33 Aaron

I don't want to hijack Derek's chem blog, and this isn't the place to discuss paleoclimatology. When you have some time, research on your own atmospheric CO2 over GEOLOGIC time, not Al Gore's beloved period since the Industrial Revolution. Look at the data over the last 500 million years. You will find that the current warming is to be expected as we have been experiencing below "normal" temps globally for about 3 million years. An interesting chart on the Wikipedia entry for "ice age" is one entitled "65 million years of climate change." It is one chart that wasn't in Gore's movie.

By the way, the problem I have with AGW is that some SCIENTISTS claim the debate is over. Imagine if Einstein had listened to those who said that the debate about gravity was over in 1900. In science, no debate is ever truly "over." If you have evidence to refute current ideas, the debate is still open.

Permalink to Comment

39. Fred Langa on September 18, 2009 8:42 AM writes...

Just to inject a note of levity, doesn't the image on the right look like a Smurf photocopied his naked ass?

Permalink to Comment

40. Terry on September 18, 2009 7:19 PM writes...

This. Is. Awesome.

Permalink to Comment

41. James on September 18, 2009 7:39 PM writes...

I am an amateur scientist at best, but I'd like to point out that most of the cross-talk on here that debunks these findings, makes specific references to all the orbital theories out there. Theories. Let me say that again. Theories.
Hence, we do not know what orbitals ACTUALLY look like. The best we have are mathematical calculations that PREDICT their structures.
Whether or not these images or results are reproduceable, since all our knowledge is based on theory, we cannot automatically say "nay" just because something doesn't assume our own suppositions.

Permalink to Comment

42. Micky on September 19, 2009 12:34 AM writes...

You are not really "seing" anything. Just a diagram of one set of probability outcomes where the fuzziness represents a probability zone.

It's an illusion based on one aspect of "fact".

Electrons are not tiny dots of "stuff" that materialise in one area, they are actually not there at all, as dots of hardness. They are consistent patterns of interaction that we don't yet understand the underlying nature of.

They certainly do follow "laws" and have properties that we equate to such things as "energy" or mass. But as to what all this stuff is (which can appear out of seeming nothing and then happily disappear") we have yet to find out.

Cheese would be my bet, but then I am a mouse.

Permalink to Comment

43. Nick K on September 19, 2009 10:33 AM writes...

I must be missing something here: how can they "isolate" a p orbital? Surely an isolated carbon atom looks like an accumulation of 1s, 2s and 2p orbitals (or their spatial electron densities) all centered on the nucleus?

Permalink to Comment

44. Nick K on September 19, 2009 10:34 AM writes...

I must be missing something here: how can they "isolate" a p orbital? Surely an isolated carbon atom looks like an accumulation of 1s, 2s and 2p orbitals (or their spatial electron densities) all centered on the nucleus?

Permalink to Comment

45. Hap on September 19, 2009 12:45 PM writes...

1) I assume (but don't know) that the instrument is looking at single energy of emission, so that it can only see the electrons in the orbitals which can absorb it (so at one energy, you see the electrons in the s-like orbital, while at another you see electrons in a p-like orbital). That might explain why you only see one orbital's e density at a time.

2) These are theories, but they have to explain a large body of data - they aren't drunken suppositions made after a bottle of Popov in cheap Kool-Aid. They still have to deal with new information, but it isn't like there wasn't a lot of old information that the theories had to account for.

Permalink to Comment

46. mivadar on September 22, 2009 10:45 AM writes...

Some notes:

The carbon atom they imaged is at the end of a carbon nanotube, attached to a graphene surface. i.e. There are no hydrogen atoms or any other species involved - the emitting atom is at the end of a (pure) carbon chain.

The idea of field emission microscopy is to have a conductive surface (tip), and subject it to sufficient voltage to induce the field emission of electrons, which are then caught on an electron-sensitive surface.
The voltage applied determines which electrons are emitted (corresponding to the work function), and the spatial configuration of the emitted electrons is magnified by them travelling along the field-lines and hitting the electron-sensitive surface a macroscopic distance away.
(You basically get a "blown up" image from where electrons with a certain work function came from.)

This is usually used to image crystal planes, or atomic positions within a crystal lattice. The first brilliant thing in this paper is that they managed to create a conductive tip that they attached to a FEEMS surface which is one atom wide.
(i.e. a nanotube "wire" ending in a single atom, rather than a tip ending in a certain crystal plane).

The second brilliant thing is that they have an electron-sensitive surface sensitive enough that it can detect field emission from a single orbital, an appreciable distance away.

The image is not created "of the orbital" in the sense that it's not a snapshot of a single electron.
It's a cumulative image of single-electron emission events from the tip carbon atom, at an emission energy corresponding to an s- or p-orbital. The electrons are "re-supplied" (after the individual emissions) from the conductive nanotube (attached to the graphene surface attached to the electrode).
The image created is, again, a magnified representation of where the electrons came from, so the cumulative image will correspond to their residence probability at any one point at the time of emission.

This probability (square of the wave function) looks like the pretty pictures in text books.
It's already amazing they managed to reproduce the forbidden region in p orbitals - I wouldn't worry too much about the symmetry (or lack thereof) in s orbitals.

As to the single p-orbital - applying a transverse magnetic field takes care of that.

By the way, it's blue for the same reason every AFM figure in publications is yellow/orange.
Greyscale doesn't tend to come out nicely on screen, so people tend to apply a colour-saturation scale in stead.
(About the AFM - the standard colour setting in the AFM software of the company most used in western labs is yellow/orange ... they have a load of other colour-schemes, but almost no one bothers to change it.)

Permalink to Comment

47. undergrad on September 22, 2009 6:10 PM writes...

Bye the bye, who came up with those structure of orbitals that we see in the standard text books, then? If we are seeing pictures of orbitals just now, that must have been some imaginary work's of some genius to come up with the shapes back then. I am curious to know who was it that proposed those structures to the orbitals? If the this paper is true, and the above explanations in the comments are true, then the experiments of Rutherford and others also should have seen some pattern of it (orbitals) when they emitted electrons, and neutrons and protons etc. If Schrödinger wave theory is applied...electron is a wave, and how can you picture a wave. I am having hard time to understand all this. Can somebody help a kid ho wants to go to medical school, but who is imposed to take OCHM....if the shit of atomic structure is not clear to me, how the things are going to be understood at molecular level, and supramolecular level like DNA and RNA. To me it is only clear that there is no bullshit like orbital like 's' or p or d or f or sp3 for that matter. What is out there is space around nucleus which is shared by electrons equally and oriented in 3'dimensional space so as to alleviate any repulsions between them and yet be in the vicinity of the nucleus. Then how do we explain the number of bonds.....only as many as required to that of noble gas...would participate in bonding. There is no bull shit like inner core and valence core. Why is that 8 magic number. May be if you have as many as those of noble gases, they could be arranged in the atom , and if anything is short then it is unstable and will try to me MO makes more sense. However, when you combine all the electrons and neutrons of atoms in to molecules, you should be able to make Helium from hydrogen under lots of external force to overcome the doesn't make sense either. I am lost!

Permalink to Comment

48. Goblue on September 23, 2009 2:56 PM writes...


The Schrodinger wave function has certain eigenvalues (psi values) that provide satisfactory solutions requisite operator. Psi squared (sqaure of the function) then gives rise to the s, p, d orbitals etc that we draw. As psi squared gives a probability density, the surface is usually drawn to encompass the volume in which the electron can be found 90% of the time. So in essence this allows that nicely drawn picture to cover the fuzziness that electrons would really exhibit.

The key though is that previously the orbitals were only predicted mathematically, but that theory reliably explained much of the observable chemical properties and reactivity. Thus it was accepted. This image just provides additional support for the previous theory.

Permalink to Comment

49. seb on September 24, 2009 2:29 AM writes...

#46. Mivadar. You're scary good. Could you tear apart some of the chemistry on this blog:


Like, what if experiments are required to calibrate the experimental conditions to maximize yield and purity, but, you don't get to conduct any experiments? what's the most you can say for sure with zero experiments? What theories rely the most on experiments to become utilized? How about the Arrhenius theory of temperature dependence such that "rate" is not blindly maximized. A little bit slower rate is better than lower yield or by-products. Do reactions come with infinitely variable rates?

Permalink to Comment

50. design on September 24, 2009 6:30 AM writes...

The comments here are either...

1) Really smart people who refute what we are actually seeing
2) Total dipshits who dont care but comment to make a joke anyways

Permalink to Comment

51. Jose on September 24, 2009 6:56 AM writes...

Wow! Seb's link above in worth a (short) perusal. Complete crackhead-lithium-salt Donald Crownhurt-esque insanity. Eg, "Cuz the hydrogen state combined with heat, creates a pressurized vacuum state."

Permalink to Comment

52. CH time on October 26, 2009 10:41 PM writes...

This is extra silly. A human's eye can only detect wavelengths in the visible light spectrum, we could NEVER see an orbital because the diameter of an atom is much smaller than 400 nm. The diameter of a typical atom is in the x-ray spectrum.

Permalink to Comment

53. Tom on May 29, 2013 2:25 PM writes...

@CH time - you make a very sound point. The next time a police helicopter finds me hiding at night in some bushes I will be careful to explain to them that they must have cheated since they could NEVER have seen me in infrared frequencies. For that they'd need some preposterous handwavium such as a non-human-optic technology (let's call it FLIR) and a false-colour rendering for ease of interpretation.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry